Practical method to facilitate estimation of Streptococcus mutans levels in saliva

0095-1137/79/05-0584/05$02.00/0

Practical

Method to Facilitate Estimation of

Streptococcus

mutans

Levels

in

Saliva

B. KOHLER* AND D.BRATTHALLt

Departmentof Cariology, Facultyof Odontology, University of Goteborg, Goteborg 33,Sweden

Received forpublication5March1979

A methodwasdeveloped tofacilitatethe estimation ofStreptococcusmutans levels in saliva. Saliva-contaminated wooden spatulas were pressed directly against an elevated agar platecontaining a selectivemedium. The results were comparedwith the number of S. mutansper1mlofparaffin-stimulatedsaliva. It was shown that the spatula method gave a good estimation of the level ofS. mutansinfection. The incubation was also made inexpiredair instead of 95% N2-5% C02. The outgrowth was in good agreement with that after conventional incubation. The methodis useful inepidemiologicalstudies or inselecting persons

at a high caries risk, and when ordinary saliva sampling cannot be done, for

example in small children. Compared with conventional saliva

sampling,

this method

requires

less time and materialat

sampling

aswell as atthe

laboratory.

Several

studies have

reported

a

positive

cor-relation

between dental caries and the

degree

of infection with

Streptococcus

mutans

(4, 9, 10).

There is also some evidence thatan increased

level

of S.mutansmay

precede

the

development

of caries lesions

(6,

9). Therefore,

someinterest hasbeen

generated

inthe

feasibility

of

including

this

microorganism

in caries

prediction

tests. Recent data also suggest that

analysis

of the number of S.mutansin saliva is

superior

in the

prediction

of caries than Lactobacillus counts,

amount of dental

plaque,

saliva secretion rate,

and

salivary

buffering capacity (7),

at least in children of

approximately

10yearsofage.

As

pointed

out

by Westergren

and Krasse (13), current

microbiological knowledge

is

rarely

applied

in

general

dental

practice.

This situation may be due in part to the reason that

simple

methods for

bacteriological diagnosis

are lack-ing. For

example,

analyses

ofS. mutans

usually

involve several steps suchas

collecting

asaliva or a dental

plaque

sample, transferring

it to a transport medium, and transporting it to a lab-oratory for processing, including dilution, plat-ing, and the anaerobic incubation of several agar plates for each sample.

Although

a

recently

published method involv-ing the use of a micropipette (13) has simplified the

procedure

substantially, there is still a need forfurtherimprovements. In this paper, we pre-sent a method by which several of the steps mentioned above can be excluded. In addition, themethod can be used for

small

children

from

tPresent address: Department of Cariology, Faculty of Odontology, UniversityofLund,21421Malmo,Sweden.

whom saliva can be collected only with great

difficulty.

It is basedon our earlier observation (8) that persons with different degrees of S. mutansinfection transfera

proportional

number of this

organism

when

they

contaminateaspoon with saliva.

MATERIALS AND METHODS

The method involvesbrieflythefollowingsequence

ofsteps:(i)thesampleiscollectedby placingawooden

spatula in the mouthinordertowetit withsaliva;

(il)

thespatula is pressed directly againstaselectiveagar

plate; (iii) plates are incubated anaerobically or in

sealed plastic bags containing expired air; (iv) the

number of bacterial colonieson apredetermined

sur-face oftheplatesisestimated.

In thispaperthe results of the methodhave been

compared with the results of conventional

paraffin-stimulated saliva samples. The spatula method was

performed twice: immediately before and after the

paraffin chewing.

Afirst series ofexperiments included37 adult

pa-tients.Sampling ofbacteria fromthe oral cavitywas

performed witha1.8-mm-wide wooden spatula (Fig.

1). About 3 cm of thespatulawas introduced intothe mouth and turned around 10 times to contaminate it

with saliva. When thespatula was taken out ofthe

mouth,any excess of saliva was wiped off against the

lips.Each side of the spatula was then pressed against

disposable contactpetri dish (Nunc, Roskilde,

Den-mark;similardishes are named RODAC plates in the

U.S.)with anelevatedlevelof theselectiveagarMSB

(mitissalivarius agar with 15% sucrose and 3.3 mgof

bacitracin,selective for S.mutans, per liter) (5) (Fig.

1). The agarplateswereincubated at 37°C for 48 h in 95% N2-5%C02.

Theparaffin-stimulatedsalivasample was then col-lected. Each subject chewed a piece of paraffin wax

584

on February 7, 2020 by guest

http://jcm.asm.org/

ESTIMATION OF S. MUTANS LEVELS IN SALIVA 585

(~2 g) for about 2 min. Saliva was collected at the

sametime,and 1 ml of the sample was then transferred

to 1 ml of reduced transport fluid consisting of a

FiG. 1. MSB agar dish with an elevated level

of

agarandawooden

spatula

(left).

Sealed

plastic bag

containing

a MSB agar dish in

expired

air

(right).

balanced mineral salt solution poised with

dithiothre-itol and containing ethylenediaminetetraacetic acid

(RTF) (12) and brought to the laboratory for

prepa-ration within 3 h. Thenumber of S. mutans

colony-forming units (CFU) per1mlofsaliva wasestimated

by themicropipette method described by Westergren

and Krasse(13).

Immediately after the paraffin-stimulated saliva

sample was collected, the sampling with the spatula

method was repeated. The plates were incubatedas

mentioned above, and the number of colonies

resem-bling S. mutans on a predetermined area of the tip

(approximately 1.5cm2) were counted for each side pressed against the MSB agar (Fig. 2). The mean of the two observations for each sample was compared

with theresults of the conventional saliva samples. To

obtain an estimation of the reproducibility of the

results, the number of colonies on one side of the spatula was compared with the number on the other

side. In this comparison, samples of 115subjects were

included.

Thespatula method, with a fewmodifications,was

also used for a group of children, 3 to 6 years old,

attending apublic dental clinic inGoteborg. Instead

of turning the spatula around, the children sucked a coupleof times oneach side of the spatula until it was

-A

-,,, e

A

4b.j.

P.o

:1''-.

-fi~~~~~~~-FIG. 2. Spatula samplescollectedfrom patientswith various levels of S. mutans in saliva:(A) 105,(B)6x 105,(C) 106,

(D)

9x106S.mutanspermlofsaliva.

B

D

VOL. 9,1979

on February 7, 2020 by guest

http://jcm.asm.org/

moistened. For stimulation, a small piece of paraffin

wax(-0.7 g) was then used. Depending on the child's

abilitytocooperate,0.1 to 0.5mlofparaffin-stimulated

salivawas collected. Again, the spatula method was

usedbothbefore and after the paraffin chewing. The

preparations of the sampleswereperformed as

men-tioned above. Eighty-six children participated in this

part of thestudy.

Tofind out ifS. mutans could be cultivated even

moresimply by theelimination ofanaerobic jars and

gas tanks usedfor routine cultureprocedures,

saliva-contaminated MSB agar plateswereplacedinplastic

bags (Plasteril folie, Plate Bonn GMBH, Bonn,

Ger-many) (Fig. 1). The opening of the bag was then

almost closed by welding (Impulse-welding apparatus,

Elwis, Gentofte, Denmark). The bag was then filled

with expired air from one staff member and

immedi-ately sealed completely. The bag was placed at 37°C

for48h. Asa control,duplicate samples collected at

the same time were incubated in the conventional

way.This studyincludedsamples from 54 subjects.

RESULTS

In

Tables

1

and

2the results

from

the

sam-pling

ofadults are presented. Table 1 shows the

comparison

between the

paraffin-stimulated

sa-liva sample

and the

spatula sample

takenafter

paraffm stimulation. When S.

mutans was de-tected in

the

saliva sample it was also detected

by

the

spatula method,

with one exception. In this case,however,

the saliva

count wasless than 10,000

S.

mutans per ml of

saliva.

Increased

numbers of

S.

mutans per 1 ml of saliva also

corresponded with increased

numbers

trans-TABLE 1. Comparison of S. mutans numbers in

salivaof adults after paraffin stimulation

No.of salivasamples having S. mu-S.mutans trans- tans(CFU/ml) numbers:

ferred from

spat-ula'(CFU)

O-105

101_106

>106

0-20 9 4

21-100 11

>100 1 12

a Counts transferred from ca. 1.5 cm2 of the wooden

spatulatoMSB agar.

TABLE 2. ComparisonofS. mutans numbers in salivaof adults

No. of salivasamples havingS.

mu-S.mutans trans- tans (CFU/ml) numbers: ferred fromspatula

(CFU)

O-105

105_106 >106

0-20 9 il 2

21-100 5 6

>100 4

aSpatula

samples

weretaken

before

paraffin

stim-ulation; numbersrepresent CFUtransferred from ca.

1.5cm2 ofthewoodenspatulato MSBagar.

Conven-tionalsalivasamplesweretaken afterparaffin

stimu-lation.

ferred by the spatula. Thus when the saliva

countswere morethan 106 S. mutans, the

num-ber transferred by the spatula was more than

100CFU.

In Table 2 theresultswith thespatulamethod taken beforeparaffinstimulationarecompared with the number of S. mutansper milliliter of

paraffin-stimulated

saliva. S. mutans was de-tected with the spatula method in all but six cases. Four of theserepresentedverylow saliva levels of S. mutans (lessthan 4 x 104permlof saliva).

Theresults fromthesamplingof thechildren, presented in Tables 3 and 4, show the same

pattern as for the adults. This group also

in-cluded a large number of children with unde-tectable levelsofS.mutans. Table3 shows that in ninecasesS.mutanswasnotdetectedbythe spatula sampling after stimulation. However, sevenofthese representedsaliva counts of less than 10,000 S. mutansper mlof saliva. On the other hand, in two cases low numbers of S.

mutans were detected by the spatula method

butnotinthesalivasample.Thespatula sample takenbeforeparaffin stimulation (Table 4) failed to detect S. mutans in 15 children. Thesewere also foundtohave low saliva counts.

The numberof S.mutanstransferredbyeach sideof thespatulawasfound to be in thesame

TABLE 3. ComparisonofS.mutansnumbers in

salivaof childrenafterparaffinstimulation.

S.mutans No. of salivasamples havingS.mutans transferred (CFU/ml)numbers: fromspatula

(CFU) Ob 1-105 105i016 >106

0 40 9

1-20 2 13

21-100 2 6

>100 3 11

aSee

footnote

a,Table1.

bNo colonies (0) meansless than400

CFU/ml

of

saliva.

TABLE 4. Comparisonof S. mutans numbers in

salivaof childrena

S.mutantstrans- No. ofsalivasamples havingS.mutans ferredfromspat- (CFU/mi)numbers:

ulab (CFU) 0 - 105_106 >106

0 40 15

1-20 2 8 7 5

21-100 1 1 4

>100 1 2

aThespatula samples were

collected before

paraffin

stimulation.

bSee

footnote

a,Table 1.

'No colonies (0) means less than 400 CFU/ml of

saliva.

on February 7, 2020 by guest

http://jcm.asm.org/

ESTIMATION OF S. MUTANS LEVELS IN SALIVA 587

magnitude

in 84%

of

the

samples.

In 15% of the

samples the difference

was one group

of

magni-tude.

The numbers of

S.

mutans

CFU

recovered by the two incubation methods are compared in

Table

5. In

85% of the

samples

the numbers

found by the

two

methods

were

within the

same range.

The

conventional way of incubation gave a

higher number of S.

mutans in

only

9% of the

samples.

DISCUSSION

The

numbers of

S.

mutans

colonies

obtained

with the

spatula method after paraffin chewing

were

closely related

tothe numbers of CFU in

the

stimulated saliva samples (Tables

1 and 3).

This indicates

that the

newmethod

reflects

the

salivary levels of S.

mutans.

Saliva

samples

in-stead of dental

plaque samples have sometimes

been

used for estimation of the degree of

infec-tion with

S.

mutans.

The

reason

for this is

the

localized

way in

which S.

mutans

colonies the

teeth

(1,

3,

11).

Plaque samples from different

teeth

may

thus

show

very

large variations

in

their

contents

of bacteria.

High

numbers of S.

mutans in a

saliva

sample,

on

the other

hand,

indicate

either that several teeth

are

infected

or

that

fewer surfaces

aremore

heavily infected.

In

each

case,

however, the subjects themselves

can be

regarded

as

heavily

infected.

Paraffin

chewing increases

the

shedding of

bacteria from the teeth.

A

comparison of the

spatula method

performed

before

chewing with

a

paraffin-stimulated

saliva

sample

is

therefore

not

quite

correct.

It

was

included

in

the

present

study

tosee

whether the

spatula

method

could

be used also for

patients

who

cannot

chew

par-affin

readily,

for

example

very young

children.

Although

an

unstimulated

sample

seems to

be

useful for the determination if

an

infection is

present or not, it

does

not

always

reflect the

release of bacteria

at

chewing.

TABLE 5. Comparisonof S.mutansnumbers

detected afteranaerobic andexpired-air incubation

No.b ofanaerobicallyincubatedsamples S. mutans CFUb: havingS. mutansnumbers:

expiredair

0-20 21-100 >100

0-20 20 2

21-100 2 13 3

>100 1 13

aAnaerobic incubationwasinjars filled with 95%

N2-5% CO2; expired-airincubationwasin sealed

plas-ticbags filled withexpiredair. 95%confidencelimits

for the mean difference between the two culturing

methods:2+6.

'The mean of the numberof colonies transferred

by thetwosides ofthespatula.

We

have found the

spatula method

very

sim-ple

to

handle. It is fast, and

no transport media or

dilution

stepsare necessary.

These facts

make it

convenient

touse in epidemiological studies or in

dental practice.

To further minimize the

equipment

needed if

samples

are processed

out-side

a

microbiology laboratory, the

experiments

using incubation in

an

atmosphere

of expired air

instead of laboratory

gas

mixtures

were

per-formed. S.

mutans

recoveries

were

similar

with

the

two

methods.

In

contrast,

control plates incubated

aerobically

without expired air

showed

no

growth

at

all

or

considerably fewer

colonies.

Byusing the spatula method, highly infected

patients

can

be

quickly

detected by judging the

density of the outgrowth (see Fig.

2). Patients

with

no orvery

low

levels,

below

104

CFU/ml,

of

S.

mutansare

also

easily

identified. The method

thus

seems to

be sufficient for

finding subjects

who

can act as

possible

sources

for transmission

of

S.

mutans

infection. Considering

the findings

by Klock (7), it

may

also be used for

identifying

patients with increased risk for caries

as

weil

as

for

evaluation of caries-preventive

measures.At

the

Department of

Cariology

in

Goteborg,

usu-ally

a

level of

106

S.

mutans or more per 1ml

of

saliva is considered

very

high and justifies the

use

of intensive therapy

to

decrease the number.

The

lowest limit of

detection with the method

presented

is at

least

100

times

below this high

level.

The method described is based

on

the

highly

selective medium MSB. Some of its properties

have been

studied earlier

at our

laboratory (2).

It

can

be mentioned here that the

problem with

outgrowth of

species

other than

S.

mutans

is

little, if the

agar

plates

are

used

freshly prepared

(maximum

1

week old). When

outgrowth of

other

species

occurs,

such strains

canusually be

excluded

by studying the colonial

morphology

under

a

dissecting microscope. Colonies

with

slightly aberrant forms

can

be

detected, how-ever.

These colonies

need

further

identification,

but the number of

patients with such colonies

is

so

low that

they should

not

detract from the

advantages of the method.

ACKNOWLEDGMENTS

We thankMirja Varpio and EiraUyterwaal forhelp in collecting thesamples fromthe children and Ann-Charlott

Borjessonand Ann-BrittLundbergforskillful technical as-sistance.-The assistance of NilsBlomqvistinstatistical ques-tionshas beengreatlyappreciated.

Theinvestigationwassupported bythe Swedish Medical ResearchCouncil(Projectno.4548).

LITERATURE CITED

1. Edman,D.C.,H.J.Keene,I.L. Shklair,and K.C. Hoerman.1975.Dentalfloss forimplantationand

sam-plingofStreptococcusmutansfromapproximalsurfaces of humanteeth. Arch.Oral Biol.20:145-148.

VOL. 9,1979

on February 7, 2020 by guest

http://jcm.asm.org/

588 KOHLER AND

BRATTHALL

2. Emilson, C. G., and D. Bratthall. 1975. Growth of

Streptococcus mutans onvarious selective media. J.

Clin.Microbiol. 4:95-98.

3. Gibbons,R.J., P.F. dePaola, D. M.Spinell,and Z.

Skobe.1974.Interdental localizationofStreptococcus mutansasrelated todental cariesexperience. Infect.

Immun.9:481-488.

4. Gibbons, R. J., and J. V. van Houte. 1975. Dental

caries. Annu. Rev. Med.26:121-136.

5. Gold,O.G., H. V.Jordan, and J. V. van Houte. 1973. A selective medium forStreptococcus mutans. Arch.

OralBiol.18:1357-1364.

6. Ikeda, T., H.J.Sandham,and E. L.Bradley,Jr.1973.

Changes inStreptococcus mutans andlactobacilli in plaque inrelation to theinitiationofdental caries in

Negrochildren. Arch. Oral Biol.18:556-566.

7. Klock, B. 1978.Acomparisonbetween differentmethods

forpredictionofcariesactivity, p. 17-20. In B. G. Bibby and R. J. Shern (ed.),Proceedings, Methodsofcaries

prediction.Microbiology Abstracts,specialsupplement.

Information, RetrievalInc.,Washington, D.C.

'8.Kohler,

B.,andD.Bratthall.1978.Intrafamiliallevels ofStreptococcusmutansand some aspects of the bac-terialtransmission.Scand. J. Dent. Res.86:35-42.

9. Loesche, W.J., J. Rowan, L. H.Staffon, andP.J.

Loos.

1975.TheassociationofStreptococcusmutans

withhuman dental decay. Infect. Immun.11:1252-1260.

10. Scherp,H.W.1971.Dentalcaries:prospectsfor

preven-tion.Science173:1199-1205.

il.

Shklair,

I.

L., H. J. Keene,andL.G.Simonson.1972.

DistributionandfrequencyofStreptococcusmutansin

caries-active

individuals.

J. Dent. Res.51:882. 12. Syed,S. A.,and W.J.Loesche.1972.Survivalofhuman

dentalplaque florainvarioustransport media. Appl. Microbiol. 24:638-644.

13. Westergren, G.,andB.Krasse. 1978.Evaluationofa

micromethodfor

determination

ofStreptococcus

mu-tansandLactobacillusinfection. J. Clin.Microbiol.7:

82-83.

on February 7, 2020 by guest

http://jcm.asm.org/